Saurav Jha – Email: saurav [dot] jha [at] unsw [dot] edu [dot] au

I am a third year PhD student at UNSW Sydney working on conditional generative models, and being supervised by Dr. Lina Yao and Dr. Dong Gong. Prior to my PhD, I was a research assistant at the Computer Vision Center, Barcelona where I worked with Dr. Joost van de Weijer on continual learning.

I completed my Erasmus Mundus Joint Master's Degree in Advanced Systems Dependability from the University of St Andrews, UK, and l'Université de Lorraine, France. During my master's, I interned with the MULTISPEECH group at Inria Nancy where I worked with Dr. Emmanuel Vincent on training domain-specific language models for Automatic Speech Recognition. Before that, I was a machine learning engineer at FactSet Research Systems Inc., India. I grew up in Eastern Nepal and earned my BS in Computer Science and Engineering from MNNIT Allahabad, India.

Experience

Tencent, Sydney, Australia (Sep 2024 - Present)
AI Research Intern

Working on multi-modal generative AI.

Sony Group Corporation, Tokyo, Japan (May 2024 - Aug 2024)
Research Scientist Intern

Worked on continual personalization of text-to-image diffusion models.

INRIA Nancy, France (Mar 2021 - Jun 2021)
Research Intern

Worked on training domain-specific language models for Automatic Speech Recognition based on machine translation.

FactSet Research Systems, Hyderabad, India (Jun 2018 - Jul 2019)
Machine Learning Engineer

Worked on Factset's named entity recognition and topic modelling services.

Publications

Please check my google scholar profile for an up-to-date list.

Pre-prints:

Jha S., Yang S., Ishii M., Zhao M., Simon C., Mirza J., Gong D., Yao L., Takahashi S., Mitsufuji Y., “Mining Your Own Secrets: Diffusion Classifier Scores for Continual Personalization of Text-to-Image Diffusion Models”, 2024. [Work done at Sony Group, Tokyo] Paper
Mirza, M. J., Zhao, M., Mao, Z., Doveh, S., Lin, W., Gavrikov, P., Dorkenwald, M., Yang, S., Jha, S., Wakaki, H., Mitsufuji, Y., Possegger, H., Feris, R., Karlinsky, L., Glass, J., “GLOV: Guided Large Language Models as Implicit Optimizers for Vision-Language Models”, 2024. Project
Jha S., Gong D., Wang X., Turner R., Yao L., “The Neural Process Family: Survey, Applications and Perspectives”, 2022. Paper

Conferences:

Jha S., Gong D., Yao L., "CLAP4CLIP: Continual Learning with Probabilistic Finetuning for Vision-Language Models", NeurIPS 2024. Paper Code
Jha S., Gong D., Zhao H., Yao L., "NPCL: Neural Processes for Uncertainty-Aware Continual Learning", NeurIPS 2023. Blog Paper Code
Li Y., Liu Z., Jha S., Cripps S., Yao L., “Distilled Reverse Attention Network for Open-world Compositional Zero-Shot Learning”, ICCV 2023. Paper

Workshops:

Joshi A., Renzella J., Bhattacharyya P., Jha S., Zhang X. “On the relevance of pre-neural approaches in natural language processing pedagogy”. ACL Workshop on Teaching NLP, 2024. Paper
Pelosin F.*, Jha S.*, Torsello A., Raducanu B., van de Weijer J. “Towards Exemplar-Free Continual Learning in Vision Transformers: an Account of Attention, Functional and Weight Regularization”. CVPR Workshop on Continual Learning, 2022. [* Equal Contribution] Paper Code [BEST RUNNER-UP PAPER!]
Jha S., Schiemer M., Ye J. “Continual learning in human activity recognition: an empirical analysis of regularization”. ICML Workshop on Continual Learning, 2020. Paper Code

Journals:

Jha S., Schiemer M., Zambonelli F., Ye J. “Continual learning in sensor-based human activity recognition: An empirical benchmark analysis”. Information Sciences, 2021. Paper Code
Ye J., Nakwijit P., Schiemer M., Jha S., Zambonelli F. “Continual Activity Recognition with Generative Adversarial Networks”. ACM Transactions on Internet of Things (TIOT), 2021. Paper
Jha S., Sudhakar A., Singh A.K. “Learning cross-lingual phonological and orthographic adaptations: a case study in improving neural machine translation between low-resource languages”. Journal of Language Modelling, 2019. Paper

Academic Services

Reviewer for: ICLR 2024, CPVR 2024, TPAMI, ACM Multimedia 2024, ICLR 2023, NeurIPS 2023
PC member for the Industry Track for Applied Research of the Web Conference 2025 (WWW 2025, Sydney, Australia).
PC member for Workshop Proposals, Conference on Information and Knowledge Management (CIKM 2023, Birmingham, UK)

Teaching at UNSW

COMP6713 (Natural Language Processing), taught by Dr. Aditya Joshi
COMP9418 (Advanced Topics in Statistical Machine Learning), taught by Dr. Gustavo Batista
ZZEN9444 (Neural Networks and Deep Learning), taught by Dr. Dong Gong
COMM5007 (Python Coding for Business), taught by Dr. Xiangyu Wang

2024

November: I will be presenting the poster for my NeurIPS 2024 paper “CLAP4CLIP: Continual Learning with Probabilistic Finetuning for Vision-Language Models” at the Sydney AI meetup. [I was awarded the TikTok-sponsored Best Student Presentation Award out of 48 posters!]
July: Presented the poster for my NeurIPS 2023 paper “NPCL: Neural Processes for Uncertainty-Aware Continual Learning” at the Eastern European Machine Learning (EEML 2024) summer school in Novi Sad, Serbia.
June: Presented the poster for my NeurIPS 2024 paper “CLAP4CLIP: Continual Learning with Probabilistic Finetuning for Vision-Language Models” at the 2nd Bayes-Duality Workshop in Tokyo.

2023

September: One paper accepted at NeurIPS 2023: “NPCL: Neural Processes for Uncertainty-Aware Continual Learning”, Saurav Jha, Dong Gong, He Zhao, Lina Yao. [Pre-print and Code coming up soon]
July: One paper accepted at ICCV 2023: “Distilled Reverse Attention Network for Open-world Compositional Zero Shot Learning” [arXiv], Yun Li, Zhe Liu, Saurav Jha, Sally Cripps, Lina Yao.

2022

September: New Pre-print: Saurav Jha, Dong Gong, Xuesong Wang, Richard E. Turner, Lina Yao, “The Neural Process Family: Survey, Applications and Perspectives.” [arXiv]
June: My paper “Towards Exemplar-Free Continual Learning in Vision Transformers” receives the Matrix-sponsored best runner-up award at the CLVision workshop, CVPR 2022 workshop on Continual Learning.
February: I have been awarded the University International Postgraduate Award (UIPA) for my PhD studies at UNSW Sydney.

2021

October: My blog post summarizing my Erasmus Mundus master’s experience has been selected to be published on Erasmus Mundus Association’s website. The link can be found here.
July: My master’s thesis “Learning Domain Specific Language Models for Automatic Speech Recognition through Machine Translation” done at INRIA has been awarded the best thesis presentation (out of 46 students across 3 universities) at the 2021 DEPEND summer school.
June: I have been awarded the 1000€ prize for the best student poster “Towards Better Voice Assistants” at the Digital Ethics4EU workshop “Digital Ethics: can we do better?”
April: Our paper “Continual Learning in Sensor-based Human Activity Recognition: an Empirical Benchmark Analysis” is accepted by Information Sciences.
March: I am an intern in the MULTISPEECH team at INRIA Nancy working on the H2020 Comprise project.

2020

November: “Continual Activity Recognition with Generative Adversarial Networks” gets accepted to be issued in the ACM Transactions on Internet of Things.
July: One paper accepted at the Workshop on Continual Learning, ICML 2020.
April: Appointed as the program representative to the EMA for the Erasmus Mundus DEPEND 2020/21 course.
March: My team from St Andrews wins the Barclays chatbot challenge at Hack the Burgh 2020, the University of Edinburgh. The project description is available here.

2019

April: I have been awarded the prestigious 49,000€ Erasmus Mundus scholarship for a two-year Master’s study at the University of St Andrews, UK and the Universite de Lorraine, France.

2018

June: I have joined the Cognitive Computing team at FactSet Research Systems as a Machine Learning Software Engineer.